Processing and analysis of remotely sensed Chlorophyll-a concentrations in the Gulf of Mexico

Eric Kouba, UTSA, eric.kouba@gmail.com (Presenter)
Hongjie Xie, UTSA, hongjie.xie@usta.edu

Studies on Chlorophyll-a concentrations in the Gulf of Mexico depend on converting a large mass of satellite sensor data into useful models, to determine when specific data values (tagged with time and location) are either normal or abnormal. Unfortunately, long term cycles and trends with actual Chl-a concentrations can be masked by large variations over short time spans, due to the combined effect of multiple, interrelated ocean processes. Remotely sensed Chl-a concentration measurements could include contributions from various sources: nominal 'baseline' values (possibly closer to observed minimum data values), river processes (sediment discharge and nitrate loads), broad area variability due to sea surface temperatures, transportation between ocean areas due to nominal winds and currents, seasonal variability over the annual cycle, strong weather patterns (hurricanes stirring up sediments and nutrients), other identifiable contributions (i.e. oil spills and dispersants), and inadvertent/incorrect contributions (i.e. additional light reflectivity in shallow waters interpreted as additional Chl-a). It is useful to identify and quantify these interrelated contributions within the Chl-a data itself, in order to properly assess when measured data values are not normal. Picking out any of these Chl-a contributions requires accessing and processing a large mass of satellite sensor data (SeaWiFS, MODIS/Aqua, MODIS/Terra, MERIS, and others). This paper presents a series of automated scripts based on available tools in Python, ArcGIS, and ENVI/IDL to combine different sensor datasets and to process massive time series of data into time series of monthly, seasonal, and yearly values (mean, standard deviation, minimum, maximum, normal, abnormal, etc.) for further analysis, information extraction, and interpretation. The Gulf of Mexico region is used to illustrate our methods and applications. The methodology is also useful for processing other datasets and for other study areas.

Presentation Type:  Poster

Session:  Global Change Impact & Vulnerability   (Tue 11:30 AM)

Associated Project(s): 

• Related Activity

Poster Location ID: 300